CHAPTER 17
J kANSLATKJN CRITICISM /S5
Translation Criticism
INTRODUCTION
Translation criticism is an essential link between translation theory and its practice; it is also
an enjoyable and instructive exercise, particularly if you are criticising someone else's
translation or, even better, two or more translations of the same text. (See Part II, especially
Texts 10-13.) You soon become aware not only of the large LtasteareaT, but that a text may
be differently translated, depending on the preferred method of the translator. For example:
Ceite rue, cette place ressemhlent a la rue, a la place d'alors: elks ne sontpas les memesy et, les
autres, je puis avoir Isimpression quelles existent encore.
rJacques BoreL I-Adoration "
translated by N\ Denny as
Those places look as they did then, bui rhey are not the same; and as for the others. 1
have the feeling that they still exist.
The point here is not how good this is as a translation or why it was not mure closely
translated, perhaps into: This street, this square are like the street, the square of those times;
they are not the same, and as for those others, I may feel that they still exist . . .\ but why Mr
Denny wanted to make an emotional, dramatic utterance into a calm, natural statement- Thus
there are various aspects of translation criticism: you can assess the translation by its
standard of referential and pragmatic accuracy, but if this is inappropriate and rather futile,
because there is so much to 'correct', you can consider why the translator has apparently
transposed or changed the mood so drastically; whether any translator has the right to change
en mime temps immobile et comme entre . . . dam une espece d'eternite to 'unchanging and
fixed in a sort of eternity1. How far is a translator entitled to get away from the words, to
devote himself to the message, the sense, the spirit?
I think there are absolute values of accuracy and economy as well as relative values
but these absolute values (like translation) must be continually reconsidered
184
and rediscussed in various cultural contexts; they cannot be taken for granted. (This
resembles the argument for God.} Up to now, translation has mainly followed the prevailing
and sometimes the countervailing ideology of the time: thus classicism (balance, noble
expression, Pope), romanticism (richness of folk language, local colour, Tieck, Schlegel), art
for art's sake (re-creation, Dowson), scientific realism (transference, James Strachey) all to
some extent find their reflection (Niederschlag) in translation. The challenge in translation
criticism is to state your own principles categorically, but at the same time to elucidate the
translator's principles, and even the principles he is reacting against (or following). In this
sense, good translation criticism is historical, dialectical, Marxist. In proposing my own two
translation methods, 'semantic1 and 'communicative1, I tend to think of the first as absolute,
the second as relative, but I am (pathetically) aware that both methods arc to some extent
reactions to or against Nida, Nabokov, Rieu and others. Nevertheless I think there is a new
element in translation now, as it becomes a profession. The introduction of a 'scientific*
method, the testing of any hypothesis or generalisation (itself arising from translation
examples) by a series of further data or translation examples, tends not to eliminate but at
least to reduce the range of choices, the extremes of ideology in translation. At the grossest
level, the evidence of the 'group loyalty factor1 so brilliantly detected by Ivars Alksnis in
several numbers of Paratteles (Geneva), showing the variations of nationalist and sex
prejudice in a large number of published translations of novels, would, if it were widely
disseminated, make the extremes of ideology, political and even literary, more difficult,
Nida's 1964 title to this fine book Towards a Science of Translating was prophetic;
translation (and translating) is not and never will be a science, but as the discipline that treats
(behandeh) it advances, translation's scientific frame of reference will be more generally
acknowledged.
Translation criticism is an essential component in a translation course: firstly, because
it painlessly improves your competence as a translator; secondly, because it expands your
knowledge and understanding of your own and the foreign language, as well as perhaps of
the topic; thirdly, because, in presenting you with options, it will help you to sort out your
ideas about translation. As an academic discipline, translation criticism ought to be the
keystone of any course in comparative literature, or literature in translation, and a component
of any professional translation course with the appropriate text-types (e.g., legal, engineering
etc.) as an exercise for criticism and discussion.
A translation may be evaluated by various authorities (Instanzen): (a) the reviser
employed by the firm or the translation company; (b) the head of section or of the company
(this may be described as 'Quality Control', if translations are sampled; the term is at present
being overused and broadened); (c) the client; (d) the professional critic of a translation or
the teacher marking one; and (e) finally by the readership of the published work. Ironically,
as Nabokov pointed out, many reviewers of translated books neither know the original work
nor the foreign language, and judge a translation on its smoothness, naturalness, easy flow,
readability and absence of interference, which are often false standards. Why should
186 PRINCIPLES
a translation not sometimes read like one, when the reader knows that is what it is?
Here, however, I am assuming that the evaluation, whether in the form of a critique
or a graded assessment, is done by way of a comparison between the original and
the translation. What is required at the present time is a reconsideration of many of
the translations that have most influenced indigenous cultures, of the kind that has
been signally performed by Bruno Bettelheim in his criticism of the authorised
English version of Freud's work,
PLAN OF CRITICISM
I think any comprehensive criticism of a translation has to cover five topics: (1) a
brief analysis of the SL text stressing its intention and its functional aspects; (2) the
translator's interpretation of the SL text's purpose, his translation method and the
translation's likely readership; (3) a selective but representative detailed com-
parison of the translation with the original; (4) an evaluation of the translation - (a)
in the translator's terms, (b) in the critic's terms; (5) where appropriate, an
assessment of the likely place of the translation in the target language culture or
discipline.
TEXT ANALYSIS
In your analysis of the SL text, you may include a statement of the author's
purpose, that is, the attitude he takes towards the topic; characterisation of the
readership; an indication of its category and type. You assess the quality of the
language to determine the translator's degree of licence, assuming for example that
he can reduce cliche to natural language in informative but not in authoritative
texts. You briefly state the topic or themes, but do not precis the text and do not
'plot-monger' (painfully retell the plot),
I suggest you do not discuss the author's life, other works, or general
background, unless they are referred to in the text - they may help you to under-
stand the text, but they are not likely to affect how you appreciate or assess the
translation.
THE TRANSLATORS PURPOSE
The second topic, your attempt to see the text from the point of view of this
translator, is sometimes overlooked in translation criticism. You may decide that
the translator has misinterpreted the author by omitting certain sections of the text
-notoriously, the first English translation of Hitler's Mein Kampf by Captain E. S.
Dugdale contained only about a third of the original, and omitted the most virulent
anti-semitic passages. The translator may have decided to deliberately antiquate
TRANSLATION CRITICISM 187
the narrative and/or the dialogue of his version, e.g., allora tornd - 'Eftsoons he turned1,
to moderate the figurative language of the original or to * liven up' simple sentences
with colloquial and idiomatic phrases: se tremper hdtivement dam les eaux baptismates
europeennes a Strasbourg - 'they are hastily initiated into the work of the Assembly at
Strasbourg'. Normally all translations are under-translations, less particularised than the
original, notably in its descriptive passages {elle est bien laide - 'she is as ugly as sin')
rather than its dramatic, and in its mental rather than its physical passages; you have to
establish whether the translator has attempted to counteract by over-translating,
resulting usually in a text somewhat longer than the original: // etait bien ckarpente -
*He was well built'. You have to assess to what extent the text has been deculturalised,
or transferred to the TL culture: Jeu ou genxillesse, Luque avait eie entreprenani dans
la voiture - 'Whether to be friendly or by design, Luque had not been idle in the car/ In
interpreting the translator's intention and procedures, you are here not criticising them
but attempting to understand why he has used these procedures. It is all too easy for a
reviewer to pounce on a translation's howiers, listing them one after another,
triumphantly discovering faux amis, wayward and stretched synonyms ^wistful1
translated as triste or nachdenklich), stiff and old-fashioned structures, which, in some
situations, may be perfectly natural (*Thus, by the hand of God and man, has the city
emerged Largely unblemished* - official guide to York)*, anachronistic colloquialisms,
literal translations of stock metaphors, and to ignore the fact that translators are
vulnerable, that good translations can and do tolerate a number of errors, and that
translators who translate in a stiff, old-fashioned, colloquial or racy style that does not
square with the original may be doing so deliberately, however misguidedly. If so, it is
your job as critic to suggest the reasons. (In a better world, these would be given in the
translator's preface.) In any event, here you empathise with the translator, and you
distinguish between incompetence (inadequate knowledge of SL and/or topic) and a
translation method which may be too idiomatic or too academic for your own tastes but
which appears consistent,
COMPARING THE TRANSLATION WITH THE ORIGINAL
Thirdly, you consider how the translator has solved the particular problems of the SL
text. You do not take the points successively; you group them selectively under
general heads: the title; the structure, including the paragraphing and sentence
connectives; shifty metaphors; cultural words; translationese; proper names;
neologisms; 'untranslatable' words; ambiguity; level of language; and, where relevant,
meta-language, puns, sound-effect.
This third section of your critique should consist of a discussion of trans-
'Noteihc French translation: Ainsi, graced la Ttaiure [ sic] el a ux hammes, lavilleettp<nveTtueau20esticte
en grande partie uuacte. The translator's ideological interference is perhaps rather more convincing [hart
the English original.
188 PRINCIPLES
lation problems and not quick recipes for a "correct1 or a better translation- Why, for instance,
did the translator within the contest prefer 'less intensely1 to 'less acutely1 or 'with less
intensify' for vivre avec moins d'acuite} Why did he prefer 'uncharted territory* to lterra
ignota* for terra ignoia'? (Latin tags more familar to French than to English educated
readers?) Why was 'drastic statement1 preferred to 'severe judgment' for jugement severe? (It
can be justified on the ground that French has no obvious one-to-one translations for 'drastic'
or 'statement', and therefore the translator was merely exploiting French lexical gaps; further,
jugemeni has a wider semantic range than 'judgment', which would be rather heavy in this
context,)
This third section is the heart of the critique; normally it has to be selective since, in
principle, any passage that diverges from literal translation in grammar, lexis or 'marked'
wrord order (as well as any deliberate sound-effect) constitutes a problem, offers choices,
requires you to justify your preferred solution. Why was Un historien contemporain ecrivait^
ily a quelques annees, que , . . changed to 'Some years ago it was remarked by a
contemporary historian that . - ,' instead of 'A contemporary historian stated, a few years ago,
that. . .'? Clearly 'Some years ago1 is a more natural, less marked, word order when placed at
the head of the sentence rather than in parenthesis, but there seems no good reason for
passivising the sentence and replacing ecrivait with 'remarked'.
THE EVALUATION OF THE TRANSLATION
Fourthly, you assess the referential and pragmatic accuracy of the translation by the tranla
tor's standards. If the translation is not a ciear version of the original, you consider first
whether the essential 'invariant' element of the text which consists usually (not always) of its
facts or its ideas is adequately represented. However, if the purpose of the text is to sell
something, to persuade* to prohibit, to express feeling through the facts and the ideas, to
please or to instruct, then this purpose is the keystone of the invariance, which changes from
text to text; and this is why any general theory of translation invariance is futile, and I am at
least a little sceptical about making a rule of Tytler's 'the complete transcript of the ideas of
the original work precedes style and manner of writing1 or Nida's 'form is secondary- to
content' (though I accept that form in translation must be changed to accommodate meaning)
given that the keystone of invariance may be expressed as much through words of quality
(adjectives, concept-words, and degree) as through words of object and action.
After considering whether the translation is successful in its own terms, you evaluate
it by your own standards of referential and pragmatic accuracy. Yci have to avoid criticising
the translator for ignoring translation principles that we.e not established nor even imagined
when he was translating. The main question here is the quality and extent of the semantic
deficit in the translation, and whether it is inevitable or due to the translator's deficiencies.
Further, you assess the translation
TRANSLATION CRITICISM 189
also as a piece of writing, independently of its original: if this is an 'anonymous 4
non-individual text, informative or persuasive, you expect it to be written in a natural manner
- neat, elegant and agreeable. If the text is personal and authoritative, you have to assess how
well the translator has captured the idiolect of the original, no matter whether it is cliehed,
natural or innovative,
THE TRANSLATION S FUTURE
Finally, in the case of a serious text, say a novel, a poem, or an important book* you assess
the work's potential importance within the target language culture. Was it in fact worth
translating? What kind of influence will it have on the language, the literature, the ideas in its
new miheu? These questions should, in my opinion, be answered in the translator's preface,
but the tradition of the translator's anonymity dies hard. This is the translation critic's attempt
to 'place' the translation in its unfamiliar surroundings.
MARKING A TRANSLATION
1 close this chapter with some observations about the difficulties of assessing a translated
text. The above scheme has illustrated two possible approaches, the functional and the
analytical. The functional is a general approach, the attempt to assess whether the translator
has achieved what he attempted to do and where he fell short. This response is in terms of
ideas. Details tend to get missed out. To some extent this is a subjective approach, the
equivalent, in the case of a teacher grading a script, of 'impression marking', and therefore
unreliable.
The analytical approach is detailed. As I see it, it rests on the assumption that a text
can be assessed in sections and that just as a bad translation is easier to recognise than a good
one, so a mistake is easier to identify than a correct or a felicitous answer. I assume that all
translation is partly science, partly craft, partly art, partly a matter of taste. Firstly, science.
'Science1 here is a matter of wrong rather than right, and there are two types of 'scientific'
mistakes, referential or linguistic. Referential mistakes are about facts, the real world,
propositions not words. Statements like 'water is air1, 'water is black', 'water breathes', etc.
are referential mistakes {though as metaphors they may be profoundly true). Referential
mistakes exist in 'fiction' (i.e., creative literature) only when it incorrectly depicts the real
world now or in history. They reveal the ignorance of the translator, or worse, of the writer,
which the translator has 'copied'. Linguistic mistakes show the translator's ignorance of the
foreign language: they may be grammatical or lexical, including words, collocations or
idioms.
Referential and linguistic mistakes are marked (or regarded) negatively - a figure
deducted from a total for a sentence or a paragraph, or as part of a total deficit. In the real
world, referential errors are both more important and potentially
n -G-
190 PRINCIPLES
more dangerous than linguistic errors, although both in the educational system (many
teachers) and amongst laymen they are often ignored or excused - 'after all, that's what
the original says, the translator's job is to reproduce it faithfully*. This is misguided- A
Dutch translator once told me he was paid three times his normal rate for a translation
he never did - he simply pointed out to his client that the (financial) text was full of
dangerous errors.
Secondly, translation is a craft or skill. The skill element is the ability to follow
or deviate from the appropriate natural usage: pragmatic and persuasive in vocative
texts, neat in informative texts, hugging the style of the original in expressive and
authoritative texts - you have to distinguish 'right1 from odd usage, to gauge degrees of
acceptability within a context. You can say lat present the railways are working on
improving their computer links', and whilst you will never get that precise nuance of
informality and continuous effort of 'working on' in another language, you will get a
servicable equivalent, 'trying to improve'. However, mistakes of usage would be easily
identified in a sentence such as ' con temporarily /for the nonce the railroads are
operating /functioning/labouring on bettering/beautifying/embellishing their computer
liaisons/relations/ These are mistakes of usage, due firstly to an inability to write well,
secondly perhaps to misuse of dictionary, thirdly to disregard of faux amis (deceptive
cognates), fourthly to persistent seeking of one-to-one equivalents; fifthly and mainly
to lack of common sense. Where transiationese is written by a native SL translator no
one is surprised; where it is written by a native TL translator it sounds absurd but is
just as common and is due to carelessness coupled with mesmerisation with SL words
at the textual level. The idea that translators, particularly of non-literary texts (infor-
mative texts), have to write well is far from generally accepted - many believe that,
where facts are concerned, style takes second place. But the truth is, it is the style that
ensures that the facts are effectively presented - bear in mind, when I think of style, I
am not thinking of 'beauty*, 1 am thinking of the fight against expressions like the
trade unionist's 'at the end of the day* and the jargon-monger's timentionality\
'translationality* and 'integrality*. There is a certain 'plainness* (a unique,
'untranslatable' word with an exceptionally wide semantic span: 'honest, direct, smooth,
simple, clear unadorned') about ^x>d usage which makes it difficult to regard it as a
'plus' in translation. Whilst mistakes of truth and language are graver than mistakes of
usage, it is skilled usage that ensures successful transmission.
Thus far I have described negative factors in assessing translation. The third
area, translation as an art, is a positive factor. It is the 'contextual re-creation' described
by Jean Delisle, where, for the purpose of interpretation, the translator has to go
beyond the text to the sub-text; i.e., what the writer means rather than what he says, or
where, for purposes of explanation, he produces an economical exposition of a stretch
of language. When fidiliti aux different* bilans translates neatly as 'adherence to the
various schedules1; when Bntgleisung^ used figuratively, becomes 'complication*;
when, in a text on electrocardiograms, Artefaki is rightly translated by the technical
term 'artefact', i.e. an electrocardiogram wave that
TRANSLATION CRITICISM 191
arises from sources other than that of the heart, e.g., a mechanical defect; when a translator
brings out an inference or an implication a little more clearly than in the SL text (say the
literal as well as the figurative significance of a metaphor); when participation has to be
translated as the 'involvement' of the endocrine system to indicate a verb-noun's or a gerund's
missing case-partner (or 'referring consultant' is le medecin-consultant adressant les sujets a
la clinique); when a cultural word is neatly explained ('he enjoyed the bananas and meat in
his tapadas snack1); when a sound-effect or a colloquialism in one part of a clause is
compensated in another (pipe mise dans son nezy Lpipe stuck in his mouth (or gobV) - these
can be described as creative translation, a find, a happy or elegant solution. Creative
translation usually has the following features: (a) a 'surface' translation is not possible; <b)
there are a variety of solutions, and ten good translators will produce this variety; (c) the
translation is what the writer meant rather than what he wrote. The solution closest to the
original is the best pragmatically, has to be weighed against referential accuracy, and there is
no clearly superior version,
If a book on A. von Humboldt starts: Alexander von Humboldt - ja, warum denn?, a
coarse translation would suggest 'Why write a book about Alexander von Humboldt of all
people?' But since the style is more refined, one might try: it may appear strange to be
writing a book about Alexander von Humboldt\ or simply: 'Alexander von Humboldt. , , Yes,
but why?1 The first version is referentially, the second pragmatically closer. The third is brief
and closest. This is creative translation.
The fourth area of translation, that of taste, has to be accepted as a subjective factor.
This area stretches from preferences between lexical synonyms to sentences or paragraphs
that under- and over-trans late in different places, e.g., for Sa compagne offrait Vimage d'une
figure admirabiement karmonieuse one has a choice between 'His companion's face
presented a picture of admirable harmony' and lHis companion's face was a picture of
admirable harmony.' Inevitably, the critic has to allow for his own taste for or bias towards
either 'literal' or 'free* translation. The taste area, on the fuzzy perimeter of translation (with
science at its centre!, renders the concept of an ideal, perfect or correct translation a
nonsense, and is itself an essential concept; the consequence is that a sensitive evaluation of
a translation is cautious and undogmatic - usually!
You will notice that in my analytical approach to translation criticism, the negative
factors of mistakes of truth, language and usage tend to outweigh the positive factors of
creative translation the felicitous renderings that make a translation not only accurate but
effective, whether we are discussing an advertisement, or a short story. However, accuracy
can also be assessed positively, with marks given for accurate renderings of sentences or
paragraphs, and deducted for mistakes; this is called 'positive marking' and is becoming
more favoured by examination boards. Being the reverse of negative marking, it often
achieves the same result. The paradox is that items of 'creative translation' are likely to
receive less credit in positive than in negative marking; since a competent translation of a
sentence gets the maximum mark, nothing is left for a 'happy' translation.
192 PRINCIPLES

QUALITY IN TRANSLATION

The question remains: What is a good translation? What fails? (And what is a bad translator
sans emploi-, as the great Louis Jouvet phrased it inimitably in Quai des Brumes?) What is a
distinguished translation? 'Often we cannot agree what a particular translation should be like.
But can one teach what one does not know?' (Neubert, 1984, p. 69). Rhetorical questions
such as: would you employ this man to do your translations? are useful only because they
produce an immediate in s t inc t i ve react ion.
Ultimately standards are relative, however much one tries to base them on criieria
rather than norms. A good translation fulfils its intention; in an informative text, it conveys
the facts acceptably; in a vocative text, its success is measurable, at least in theory, and
therefore the effectiveness of an advertising agency translator can be shown by results; in an
authoritative or an expressive text, form is almost as important as content, there is often a
tension between the expressive and the aesthetic functions of language and therefore a
merely 'adequate' translation may be useful to explain what the text is about (cf. many
Penguin Plain Prose translations), but a good translation has to be 'distinguished' and the
translator exceptionally sensitive; for me, the exemplar is Andreas Mayor's translation of
Proust's Le Temps reirouve- * Time Regained*.
In principle, it should be easier to assess a translation than an original text, since it is
an imitation. The difficulty lies not so much in knowing or recognising what a good
Translation is n as in generalising with trite definitions that are little short of truisms, since
there are as many types of translations as there are of texts. But the fact that there is a small
element of uncertainty and subjectivity in any judgment about a translation eliminates neither
the necessity nor the usefulness of translation criticism, as an aid for raising translation
standards and for reaching more agreement about the nature of translation.